In this paper we present a first implementation of a linguistic expert system, which combines Bayesian inference and the knowledge of a domain expert, the linguist. The system is capable of automatically deriving the gradations of four text types -- descriptive/narrative, explicatory/informational, argumentative/persuasive, instructional -- from web pages. The evaluation of the system against human assessment shows that the system is effective and promising.
|Title of host publication||Proceedings of the 10th ESSLLI student session|
|Number of pages||10|
|Publication status||Published - 2005|
|Event||Proceedings of the 10th ESSLLI student session - Edinburgh, UK|
Duration: 1 Jan 2005 → …
|Conference||Proceedings of the 10th ESSLLI student session|
|Period||1/01/05 → …|